File | Open | Recording: File not found?

Trying to create a recording (planning to capture in real time, then study at leisure).
Clicked File | Open, navigated to F:… existing folder, entered Filename: [box]; pop-up error no matter what filename I type, “File not found. Check the file name and try again”

Hi @PFKlammer. A couple of easy questions:

  1. Have you confirmed that the “.jls” file exists in that location using Windows Explorer?
  2. Are you including the “.jls” extension in the file name? Windows hides extensions by defaults, but this dialog requires the extension (at least for now).

Note that given the path starting with “F:”, I presume that you are on Windows.

No, the .jls file does not exist, yet; I am trying to create a fresh recording from scratch. Am I supposed to pre-seed with an empty .jls file first?

Yes, Windows; did try also on C: drive, under same folder as Joulescope installation creates Log folder (C:\Users\Pete\AppData\Local\joulescope\log).

The Windows file dialog window has a box for filename and pulldown list button indicating “Joulescope Data (.jls)” as default file type extension; I presumed I don’t have to type the .jls in the File Name box?

Ah, I understand now. To create a new recording:

  • Click ViewOscilloscope
  • Click the record button record
  • Enter your path & filename if desired
  • Click Save
  • Record what you want, and then click the record button again.

To view the recording, click FileOpen and then select your file.

Also: GUI state strangeness after attempted File Open failure: gave up with “Cancel” from file dialog, now the Play/Stop triangle-in-circle icon button is grayed off (even though hover still displays “Enable to capture data …” – searching around menu discovered “Device” pulldown has selected “disable”, so chose “Joulescope:000123” and now Play/Stop button works again. File dialog disables Device selection?

Hmmm… with View | Oscilloscope, second icon is gray and disabled.
Ah… until I click Play (Enable) first.
So that means I can’t start recording until capturing is already underway, right?
I’m not sure that gives me control of where/when the recording starts; or maybe it does.
I guess what I was expecting, what I would prefer, is to pre-set my recording file before I start the capture session.
And, as mentioned elsewhere, the ability to remote-off/on the target: the whole point being, how to produce a recording starting with power-on of target, without many megabytes of needless lead-up prologue data recorded prior to power-on.

Hi @PFKlammer

Sounds like you are trying to do some very specific sequence of actions in a repeatable way. As I mentioned over on Power on/off relay control, you can get better controlled, unattended operation with a script. My offer to create this script still stands. Just let me know!

If you want to use the UI for this, you will have to click a few buttons and select the filenames manually.

When you open a file, the file becomes the active “Device”. On cancel, the UI currently does not keep information to go back to what it was doing. In 0.4.5, if you want to resume using your connected Joulescope, you will have to select it. I looked at the code, and I modified the UI to defer any changes until AFTER a file is successfully selected which means cancel would just not do anything to the selected device. Done and will be in 0.4.6 :wink:

If you want to record the behavior of your target device at power-on, as it happens in real-time:

  1. Select Current Range = off
  2. Press the Play button (if not already streaming live data).
  3. Press the Record button and select the filename to start recording. Immediately after you press the “Save” button, the file is created and captured data samples are written to the file.
  4. Select Current Range = (your desired current range, such as auto)
  5. Press the Record button to stop your recording.

Alternatively, if your duration of interest is less than 30 seconds, you can use the memory buffer and dual markers to export exactly your region of interest. If not already streaming live data from your Joulescope, press “Play”. Set Current Range = off, wait a moment, set Current Range to your desired range, and capture the power-on data to the buffer. When you have enough data, press Play to stop streaming the live capture data. Use Dual Markers (Right click the x-axis → AnnotationsDual Markers). Then left click on the marker, move it to the start, left click again. Repeat for the other marker to move it to the end. Then right click on a marker, select “Export Data”.

In either case, note that the exported jls file is currently truncated to the nearest multiple of 0.2 seconds.

And a little apology due here: with some RTFM, I understand the interface a little better now.

Maybe this deserves a new topic/thread … but …
Now examining a file captured over 40-50 seconds, cannot zoom or pan past 30 seconds ; i.e. with timescale x-axis reading 0 … 2 … … 26 … 28, cannot zoom out further, cannot click/drag/pan to right.

Well, that’s odd. Let’s see if I can duplicate this quickly…

My .jls file is 234 MB (!!!); can you translate that to seconds?

Recordings are about 8.6 MB/s (2 MSPS * 2 bytes/sample * 2 channels * ~7.5% overhead). 232 MB / 8.6 MB/s = 27 seconds. You shouldn’t be able to zoom out further than 28 seconds :wink:

I just captured 63.2 seconds which is 494 MB. Pan and zoom work fine, but you do have to be a little patient. The software has some optimization here, but needs more attention. The next big help will be discarding all but the most recent changes for animation. Right now, it attempts to draw every single requested change, which is way too much on pan.

Aha. F: drive must have run out of space at 234 MB … feature request therefore: something eye-catching on screen when recording disk writes fail?